In this paper, a multilayer perceptron guided encryption/decryption (STMLP) in wireless communication
has been proposed for exchange of data/information. Multilayer perceptron transmitting systems at both
ends generate an identical output bit and the network are trained based on the output which is used to
synchronize the network at both ends and thus forms a secret-key at end of synchronizations of the
networks. Weights or hidden units of the hidden layer help to form a secret session key. The plain text is
encrypted through chaining , cascaded xoring of multilayer perceptron generated session key. If size of the
final block of plain text is less than the size of the key then this block is kept unaltered. Receiver will use
identical multilayer perceptron generated session key for performing deciphering process for getting the
plain text. Parametric tests have been done and results are compared in terms of Chi-Square test, response
time in transmission with some existing classical techniques, which shows comparable results for the
proposed technique. Variation numbers of input vectors and hidden layers will increase the confusion
/diffusion of the schemeand hence increase the security. As a result variable energy based techniques may
be achieved which may be applicable devices/interface of the heterogeneous sizes of the network/device.
Secured wireless communication through simulated annealing guided traingulari...csandit
In this paper, simulated annealing guided traingularized encryption using multilayer perceptron
generated session key (SATMLP) has been proposed for secured wireless communication. Both
sender and receiver station uses identical multilayer perceptron and depending on the final
output of the both side multilayer perceptron, weights vector of hidden layer get tuned in both
ends. After this tunning step both perceptrons generates identical weight vectors which is
consider as an one time session key. In the 1st level of encryption process traingularized sub
key1 is use to encrypt the plain text. In 2nd level of encryption simulated annealing method
helps to generates sub key 2 for further encryption of 1st level generated traingularized
encrypted text. Finally multilayer perceptron generated one time session key is used to perform
3rd level of encryption of 2nd level generated encrypted text. Recipient will use same identical
multilayer perceptron guided session key for performing deciphering process for getting the
simulated annealing guided intermediate cipher text. Then using sub key 2 deciphering
technique is performed to get traingularized encrypted text. Finally sub key 1 is used to
generate the plain text. In this proposed technique session key is not transmitted over public
channel apart from few data transfers are needed for weight simulation process. Because after
synchronization process both multilayer perceptron generates identical weight vector which
acts as a session key. Parametric tests are done and results are compared in terms of Chi-
Square test, response time in transmission with some existing classical techniques, which shows
comparable results for the proposed system.
SECURED WIRELESS COMMUNICATION THROUGH SIMULATED ANNEALING GUIDED TRAINGULARI...cscpconf
In this paper, simulated annealing guided traingularized encryption using multilayer perceptron generated session key (SATMLP) has been proposed for secured wireless ommunication. Bothsender and receiver station uses identical multilayer perceptron and depending on the final output of the both side multilayer perceptron, weights vector of hidden layer get tuned in bothends. After this tunning step both perceptrons generates identical weight vectors which is consider as an one time session key. In the 1st level of encryption process traingularized sub key1 is use to encrypt the plain text. In 2nd level of encryption simulated annealing method helps to generates sub key 2 for further encryption of 1st level generated traingularized encrypted text. Finally multilayer perceptron generated one time session key is used to perform
3rd level of encryption of 2nd level generated encrypted text. Recipient will use same identical multilayer perceptron guided session key for performing deciphering process for getting the simulated annealing guided intermediate cipher text. Then using sub key 2 deciphering technique is performed to get traingularized encrypted text. Finally sub key 1 is used to generate the plain text. In this proposed technique session key is not transmitted over public channel apart from few data transfers are needed for weight simulation process. Because after synchronization process both multilayer perceptron generates identical weight vector which acts as a session key. Parametric tests are done and results are compared in terms of ChiSquare test, response time in transmission with some existing classical techniques, which shows comparable results for the proposed system.
NeuroCrypto: C++ Implementation of Neural Cryptography with Rijndael CipherSagun Man Singh Shrestha
This work is the software implementation of the concept of neural cryptography, which is a communication of two tree parity machines for agreement on a common key over a public channel. This key is utilized to encrypt a sensitive message to be transmitted over an insecure channel using Rijndael cipher. This is a new potential source for public key cryptography schemes which are not based on number theoretic functions, and have small time and memory complexities. This paper will give a brief introduction to artificial neural networks, cryptography and its types, which will help explain why the two communicating terminals converge to a common key in neural cryptography and will also cover the Rijndael (AES) cipher. This paper is intended to show that such neural key exchange protocol and AES encryption can be practically implemented in a high-level programming language viz. C++, which could be further extended in higher-level applications. Both CLI and GUI implementations of the software created using Visual C++ (.NET framework) are presented.
Here is my class on the multilayer perceptron where I look at the following:
1.- The entire backproagation algorithm based in the gradient descent
However, I am planning the tanning based in Kalman filters.
2.- The use of matrix computations to simplify the implementations.
I hope you enjoy it.
Understanding Deep Learning & Parameter Tuning with MXnet, H2o Package in RManish Saraswat
Simple guide which explains deep learning and neural network with hands on experience in R using MXnet and H2o package. It also explains gradient descent and backpropagation algorithm.
Complete tutorial: http://blog.hackerearth.com/understanding-deep-learning-parameter-tuning-with-mxnet-h2o-package-r
Secured wireless communication through simulated annealing guided traingulari...csandit
In this paper, simulated annealing guided traingularized encryption using multilayer perceptron
generated session key (SATMLP) has been proposed for secured wireless communication. Both
sender and receiver station uses identical multilayer perceptron and depending on the final
output of the both side multilayer perceptron, weights vector of hidden layer get tuned in both
ends. After this tunning step both perceptrons generates identical weight vectors which is
consider as an one time session key. In the 1st level of encryption process traingularized sub
key1 is use to encrypt the plain text. In 2nd level of encryption simulated annealing method
helps to generates sub key 2 for further encryption of 1st level generated traingularized
encrypted text. Finally multilayer perceptron generated one time session key is used to perform
3rd level of encryption of 2nd level generated encrypted text. Recipient will use same identical
multilayer perceptron guided session key for performing deciphering process for getting the
simulated annealing guided intermediate cipher text. Then using sub key 2 deciphering
technique is performed to get traingularized encrypted text. Finally sub key 1 is used to
generate the plain text. In this proposed technique session key is not transmitted over public
channel apart from few data transfers are needed for weight simulation process. Because after
synchronization process both multilayer perceptron generates identical weight vector which
acts as a session key. Parametric tests are done and results are compared in terms of Chi-
Square test, response time in transmission with some existing classical techniques, which shows
comparable results for the proposed system.
SECURED WIRELESS COMMUNICATION THROUGH SIMULATED ANNEALING GUIDED TRAINGULARI...cscpconf
In this paper, simulated annealing guided traingularized encryption using multilayer perceptron generated session key (SATMLP) has been proposed for secured wireless ommunication. Bothsender and receiver station uses identical multilayer perceptron and depending on the final output of the both side multilayer perceptron, weights vector of hidden layer get tuned in bothends. After this tunning step both perceptrons generates identical weight vectors which is consider as an one time session key. In the 1st level of encryption process traingularized sub key1 is use to encrypt the plain text. In 2nd level of encryption simulated annealing method helps to generates sub key 2 for further encryption of 1st level generated traingularized encrypted text. Finally multilayer perceptron generated one time session key is used to perform
3rd level of encryption of 2nd level generated encrypted text. Recipient will use same identical multilayer perceptron guided session key for performing deciphering process for getting the simulated annealing guided intermediate cipher text. Then using sub key 2 deciphering technique is performed to get traingularized encrypted text. Finally sub key 1 is used to generate the plain text. In this proposed technique session key is not transmitted over public channel apart from few data transfers are needed for weight simulation process. Because after synchronization process both multilayer perceptron generates identical weight vector which acts as a session key. Parametric tests are done and results are compared in terms of ChiSquare test, response time in transmission with some existing classical techniques, which shows comparable results for the proposed system.
NeuroCrypto: C++ Implementation of Neural Cryptography with Rijndael CipherSagun Man Singh Shrestha
This work is the software implementation of the concept of neural cryptography, which is a communication of two tree parity machines for agreement on a common key over a public channel. This key is utilized to encrypt a sensitive message to be transmitted over an insecure channel using Rijndael cipher. This is a new potential source for public key cryptography schemes which are not based on number theoretic functions, and have small time and memory complexities. This paper will give a brief introduction to artificial neural networks, cryptography and its types, which will help explain why the two communicating terminals converge to a common key in neural cryptography and will also cover the Rijndael (AES) cipher. This paper is intended to show that such neural key exchange protocol and AES encryption can be practically implemented in a high-level programming language viz. C++, which could be further extended in higher-level applications. Both CLI and GUI implementations of the software created using Visual C++ (.NET framework) are presented.
Here is my class on the multilayer perceptron where I look at the following:
1.- The entire backproagation algorithm based in the gradient descent
However, I am planning the tanning based in Kalman filters.
2.- The use of matrix computations to simplify the implementations.
I hope you enjoy it.
Understanding Deep Learning & Parameter Tuning with MXnet, H2o Package in RManish Saraswat
Simple guide which explains deep learning and neural network with hands on experience in R using MXnet and H2o package. It also explains gradient descent and backpropagation algorithm.
Complete tutorial: http://blog.hackerearth.com/understanding-deep-learning-parameter-tuning-with-mxnet-h2o-package-r
A New Classifier Based onRecurrent Neural Network Using Multiple Binary-Outpu...iosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Neuro genetic key based recursive modulo 2 substitution using mutated charact...ijcsity
In this paper, a neural genetic key based technique for encryption (NGKRMSMC) has been proposed
through recursive modulo
-
2 substitution using mutated character code generation for online wireless
communication of data/information.
Both sender and receive
r get synchronized based on their final output
.
The length of the key depends on the number of input and output neurons. Coordinated matching weight
vectors assist to generate chromosomes pool. Genetic secret key is obtained using fitness function, which i
s
the hamming distance between two chromosomes. Crossover and mutation are used to add elitism of
chromosomes.
At first
mutated character code table
based encryption strategy get perform on plain text.
.
Then the intermediate cipher text is yet again encry
pted through recursive positional modulo
-
2 substitution
technique to from next level encrypted text. This 2nd level intermediate cipher text is again encrypted to
form the final cipher text through chaining and cascaded xoring of neuro genetic key with the
identical
length intermediate cipher text block.
Receiver will perform same symmetric operation to get back the plain
text using identical key
With massive amounts of computational power, machines can now recognize objects and translate speech in real time. Thanks to Deep Learning, Artificial Intelligence is now getting smart. Deep Learning models attempt to mimic the activity of the neocortex. It is understood that the activity of these layers of neurons is what constitutes a brain to be able to "think". These models learn to recognize patterns in digital representations of data in a very similar sense to humans. In this survey report, we introduce the most important concepts of Deep Learning along with the state of the art models that are now widely adopted in commercial products.
An artificial neuron network (ANN) is a computational model based on the structure and functions of biological neural networks. It works on real-valued, discrete-valued and vector valued.
Web spam classification using supervised artificial neural network algorithmsaciijournal
Due to the rapid growth in technology employed by the spammers, there is a need of classifiers that are more efficient, generic and highly adaptive. Neural Network based technologies have high ability of adaption as well as generalization. As per our knowledge, very little work has been done in this field using neural network. We present this paper to fill this gap. This paper evaluates performance of three supervised learning algorithms of artificial neural network by creating classifiers for the complex problem of latest web spam pattern classification. These algorithms are Conjugate Gradient algorithm, Resilient Backpropagation learning, and Levenberg-Marquardt algorithm.
In information technology (IT), a neural network is a system of hardware and/or software patterned after the operation of neurons in the human brain. Neural networks -- also called artificial neural networks -- are a variety of deep learning technology, which also falls under the umbrella of artificial intelligence, or AI.
Deep Learning Interview Questions And Answers | AI & Deep Learning Interview ...Simplilearn
This Deep Learning interview questions and answers presentation will help you prepare for Deep Learning interviews. This presentation is ideal for both beginners as well as professionals who are appearing for Deep Learning, Machine Learning or Data Science interviews. Learn what are the most important Deep Learning interview questions and answers and know what will set you apart in the interview process.
Some of the important Deep Learning interview questions are listed below:
1. What is Deep Learning?
2. What is a Neural Network?
3. What is a Multilayer Perceptron (MLP)?
4. What is Data Normalization and why do we need it?
5. What is a Boltzmann Machine?
6. What is the role of Activation Functions in neural network?
7. What is a cost function?
8. What is Gradient Descent?
9. What do you understand by Backpropagation?
10. What is the difference between Feedforward Neural Network and Recurrent Neural Network?
11. What are some applications of Recurrent Neural Network?
12. What are Softmax and ReLU functions?
13. What are hyperparameters?
14. What will happen if learning rate is set too low or too high?
15. What is Dropout and Batch Normalization?
16. What is the difference between Batch Gradient Descent and Stochastic Gradient Descent?
17. Explain Overfitting and Underfitting and how to combat them.
18. How are weights initialized in a network?
19. What are the different layers in CNN?
20. What is Pooling in CNN and how does it work?
Simplilearn’s Deep Learning course will transform you into an expert in deep learning techniques using TensorFlow, the open-source software library designed to conduct machine learning & deep neural network research. With our deep learning course, you’ll master deep learning and TensorFlow concepts, learn to implement algorithms, build artificial neural networks and traverse layers of data abstraction to understand the power of data and prepare you for your new role as deep learning scientist.
Why Deep Learning?
It is one of the most popular software platforms used for deep learning and contains powerful tools to help you build and implement artificial neural networks.
Advancements in deep learning are being seen in smartphone applications, creating efficiencies in the power grid, driving advancements in healthcare, improving agricultural yields, and helping us find solutions to climate change.
There is booming demand for skilled deep learning engineers across a wide range of industries, making this deep learning course with TensorFlow training well-suited for professionals at the intermediate to advanced level of experience. We recommend this deep learning online course particularly for the following professionals:
1. Software engineers
2. Data scientists
3. Data analysts
4. Statisticians with an interest in deep learning
Learn more at: https//www.simplilearn.com
Fundamental, An Introduction to Neural NetworksNelson Piedra
An introduction to Neural Networks, eight edition, 1996
Authors: Ben Krose, Faculty of Mathematics & Computer Science, University of Amsterdam. Patrick wan der Smagt, Institute of Robotics and Systems Dynamics, German Aerospace Research Establishment
Keynote: Nelson Piedra, Computer Sciences School - Advanced Tech, Technical University of Loja UTPL, Ecuador.
Mobile devices, specifically smartphones, have become ubiquitous. For this reason, businesses are starting
to develop “Bring Your Own Device” policies to allow their employees to use their owned devices in the
workplace. BYOD offers many potential advantages: enhanced productivity, increased revenues, reduced
mobile costs and IT efficiencies. However, due to emerging attacks and limitations on device resources, it is
difficult to trust these devices with access to critical proprietary information. Therefore, in this paper, the
potential attacks of BYOD and taxonomy of BYOD attacks are presented. Advanced persistent threat (APT)
and malware attack are discussed in depth in this paper. Next, the proposed solution to mitigate the attacks
of BYOD is discussed. Lastly, the evaluations of the proposed solutions based on the X.800 security
architecture are presented.
In this paper, Space Time Block Code (STBC), Spatial Multiplexing (SM) and hybrid model with OFDM
are designed for Rayleigh fading channel. Combination of SM and STBC forms hybrid MIMO model. The
performances of the above mentioned models with different modulations such as Quadrature Phase Shift
Keying (QPSK) and Quadrature Amplitude Modulation (QAM) with multiple antennas are measured with
respect to BER. In this paper, it is shown that Hybrid MIMO provides low BER. Thus, in wireless
communication, hybrid model improves the data rate and link reliability.
Transmit antenna subset selection in mimo ofdm system using adaptive mutuatio...ijmnct
Multiple input multiple output techniques are considered attractive for future wireless communication
systems, due to the continuing demand for high data rates, spectral efficiency, suppress interference ability
and robustness of transmission. MIMO-OFDM is very helpful to transmit high data rate in wireless
transmission and provides good maximum system capacity by getting the advantages of both MIMO and
OFDM. The main problem in this system is that increase in number of transmit and receive antennas lead
to hardware complexity. To tackle this issue, an effective optimal transmit antenna subset selection method
is proposed in paper with the aid of Adaptive Mutation Genetic Algorithm (AGA). Here, the selection of
transmit antenna subsets are done by the adaptive mutation of Genetic Algorithm in MIMO-OFDM system.
For all the mutation points, the fitness function are evaluated and from that value, best fitness based
mutation points are chosen. After the selection of best mutation points, the mutation process is carried out,
accordingly. The implementation of proposed work is done in the working platform MATLAB and the
performance are evaluated with various selection of transmit antenna subsets. Moreover, the comparison
results between the existing GA with mutation and the proposed GA with adaptive mutation are discussed.
Hence, using the proposed work, the selection of transmit antenna with the maximum capacity is made and
which leads to the reduced hardware complexity and undisturbed data rate in the MIMO-OFDM system
At this present scenario, the demand of the system capacity is very high in wireless network. MIMO
technology is used from the last decade to provide this requirement for wireless network antenna
technology. MIMO channels are mostly used for advanced antenna array technology. But it is most
important to control the error rate with enhanced system capacity in MIMO for present-day progressive
wireless communication. This paper explores the frame error rate with respect to different path gain of
MIMO channel. This work has been done in different fading scenario and produces a comparative analysis
of MIMO on the basis of those fading models in various conditions. Here, it is to be considered that
modulation technique as QPSK to observe these comparative evaluations for different Doppler frequencies.
From the comparative analysis, minimum amount of frame error rate is viewed for Rician distribution at
LOS path Doppler shift of 0 Hz. At last, this work is concluded with a comparative bit error rate study on
the basis of singular parameters at different SNR levels to produce the system performance for uncoded
QPSK modulation.
A New Classifier Based onRecurrent Neural Network Using Multiple Binary-Outpu...iosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
Neuro genetic key based recursive modulo 2 substitution using mutated charact...ijcsity
In this paper, a neural genetic key based technique for encryption (NGKRMSMC) has been proposed
through recursive modulo
-
2 substitution using mutated character code generation for online wireless
communication of data/information.
Both sender and receive
r get synchronized based on their final output
.
The length of the key depends on the number of input and output neurons. Coordinated matching weight
vectors assist to generate chromosomes pool. Genetic secret key is obtained using fitness function, which i
s
the hamming distance between two chromosomes. Crossover and mutation are used to add elitism of
chromosomes.
At first
mutated character code table
based encryption strategy get perform on plain text.
.
Then the intermediate cipher text is yet again encry
pted through recursive positional modulo
-
2 substitution
technique to from next level encrypted text. This 2nd level intermediate cipher text is again encrypted to
form the final cipher text through chaining and cascaded xoring of neuro genetic key with the
identical
length intermediate cipher text block.
Receiver will perform same symmetric operation to get back the plain
text using identical key
With massive amounts of computational power, machines can now recognize objects and translate speech in real time. Thanks to Deep Learning, Artificial Intelligence is now getting smart. Deep Learning models attempt to mimic the activity of the neocortex. It is understood that the activity of these layers of neurons is what constitutes a brain to be able to "think". These models learn to recognize patterns in digital representations of data in a very similar sense to humans. In this survey report, we introduce the most important concepts of Deep Learning along with the state of the art models that are now widely adopted in commercial products.
An artificial neuron network (ANN) is a computational model based on the structure and functions of biological neural networks. It works on real-valued, discrete-valued and vector valued.
Web spam classification using supervised artificial neural network algorithmsaciijournal
Due to the rapid growth in technology employed by the spammers, there is a need of classifiers that are more efficient, generic and highly adaptive. Neural Network based technologies have high ability of adaption as well as generalization. As per our knowledge, very little work has been done in this field using neural network. We present this paper to fill this gap. This paper evaluates performance of three supervised learning algorithms of artificial neural network by creating classifiers for the complex problem of latest web spam pattern classification. These algorithms are Conjugate Gradient algorithm, Resilient Backpropagation learning, and Levenberg-Marquardt algorithm.
In information technology (IT), a neural network is a system of hardware and/or software patterned after the operation of neurons in the human brain. Neural networks -- also called artificial neural networks -- are a variety of deep learning technology, which also falls under the umbrella of artificial intelligence, or AI.
Deep Learning Interview Questions And Answers | AI & Deep Learning Interview ...Simplilearn
This Deep Learning interview questions and answers presentation will help you prepare for Deep Learning interviews. This presentation is ideal for both beginners as well as professionals who are appearing for Deep Learning, Machine Learning or Data Science interviews. Learn what are the most important Deep Learning interview questions and answers and know what will set you apart in the interview process.
Some of the important Deep Learning interview questions are listed below:
1. What is Deep Learning?
2. What is a Neural Network?
3. What is a Multilayer Perceptron (MLP)?
4. What is Data Normalization and why do we need it?
5. What is a Boltzmann Machine?
6. What is the role of Activation Functions in neural network?
7. What is a cost function?
8. What is Gradient Descent?
9. What do you understand by Backpropagation?
10. What is the difference between Feedforward Neural Network and Recurrent Neural Network?
11. What are some applications of Recurrent Neural Network?
12. What are Softmax and ReLU functions?
13. What are hyperparameters?
14. What will happen if learning rate is set too low or too high?
15. What is Dropout and Batch Normalization?
16. What is the difference between Batch Gradient Descent and Stochastic Gradient Descent?
17. Explain Overfitting and Underfitting and how to combat them.
18. How are weights initialized in a network?
19. What are the different layers in CNN?
20. What is Pooling in CNN and how does it work?
Simplilearn’s Deep Learning course will transform you into an expert in deep learning techniques using TensorFlow, the open-source software library designed to conduct machine learning & deep neural network research. With our deep learning course, you’ll master deep learning and TensorFlow concepts, learn to implement algorithms, build artificial neural networks and traverse layers of data abstraction to understand the power of data and prepare you for your new role as deep learning scientist.
Why Deep Learning?
It is one of the most popular software platforms used for deep learning and contains powerful tools to help you build and implement artificial neural networks.
Advancements in deep learning are being seen in smartphone applications, creating efficiencies in the power grid, driving advancements in healthcare, improving agricultural yields, and helping us find solutions to climate change.
There is booming demand for skilled deep learning engineers across a wide range of industries, making this deep learning course with TensorFlow training well-suited for professionals at the intermediate to advanced level of experience. We recommend this deep learning online course particularly for the following professionals:
1. Software engineers
2. Data scientists
3. Data analysts
4. Statisticians with an interest in deep learning
Learn more at: https//www.simplilearn.com
Fundamental, An Introduction to Neural NetworksNelson Piedra
An introduction to Neural Networks, eight edition, 1996
Authors: Ben Krose, Faculty of Mathematics & Computer Science, University of Amsterdam. Patrick wan der Smagt, Institute of Robotics and Systems Dynamics, German Aerospace Research Establishment
Keynote: Nelson Piedra, Computer Sciences School - Advanced Tech, Technical University of Loja UTPL, Ecuador.
Mobile devices, specifically smartphones, have become ubiquitous. For this reason, businesses are starting
to develop “Bring Your Own Device” policies to allow their employees to use their owned devices in the
workplace. BYOD offers many potential advantages: enhanced productivity, increased revenues, reduced
mobile costs and IT efficiencies. However, due to emerging attacks and limitations on device resources, it is
difficult to trust these devices with access to critical proprietary information. Therefore, in this paper, the
potential attacks of BYOD and taxonomy of BYOD attacks are presented. Advanced persistent threat (APT)
and malware attack are discussed in depth in this paper. Next, the proposed solution to mitigate the attacks
of BYOD is discussed. Lastly, the evaluations of the proposed solutions based on the X.800 security
architecture are presented.
In this paper, Space Time Block Code (STBC), Spatial Multiplexing (SM) and hybrid model with OFDM
are designed for Rayleigh fading channel. Combination of SM and STBC forms hybrid MIMO model. The
performances of the above mentioned models with different modulations such as Quadrature Phase Shift
Keying (QPSK) and Quadrature Amplitude Modulation (QAM) with multiple antennas are measured with
respect to BER. In this paper, it is shown that Hybrid MIMO provides low BER. Thus, in wireless
communication, hybrid model improves the data rate and link reliability.
Transmit antenna subset selection in mimo ofdm system using adaptive mutuatio...ijmnct
Multiple input multiple output techniques are considered attractive for future wireless communication
systems, due to the continuing demand for high data rates, spectral efficiency, suppress interference ability
and robustness of transmission. MIMO-OFDM is very helpful to transmit high data rate in wireless
transmission and provides good maximum system capacity by getting the advantages of both MIMO and
OFDM. The main problem in this system is that increase in number of transmit and receive antennas lead
to hardware complexity. To tackle this issue, an effective optimal transmit antenna subset selection method
is proposed in paper with the aid of Adaptive Mutation Genetic Algorithm (AGA). Here, the selection of
transmit antenna subsets are done by the adaptive mutation of Genetic Algorithm in MIMO-OFDM system.
For all the mutation points, the fitness function are evaluated and from that value, best fitness based
mutation points are chosen. After the selection of best mutation points, the mutation process is carried out,
accordingly. The implementation of proposed work is done in the working platform MATLAB and the
performance are evaluated with various selection of transmit antenna subsets. Moreover, the comparison
results between the existing GA with mutation and the proposed GA with adaptive mutation are discussed.
Hence, using the proposed work, the selection of transmit antenna with the maximum capacity is made and
which leads to the reduced hardware complexity and undisturbed data rate in the MIMO-OFDM system
At this present scenario, the demand of the system capacity is very high in wireless network. MIMO
technology is used from the last decade to provide this requirement for wireless network antenna
technology. MIMO channels are mostly used for advanced antenna array technology. But it is most
important to control the error rate with enhanced system capacity in MIMO for present-day progressive
wireless communication. This paper explores the frame error rate with respect to different path gain of
MIMO channel. This work has been done in different fading scenario and produces a comparative analysis
of MIMO on the basis of those fading models in various conditions. Here, it is to be considered that
modulation technique as QPSK to observe these comparative evaluations for different Doppler frequencies.
From the comparative analysis, minimum amount of frame error rate is viewed for Rician distribution at
LOS path Doppler shift of 0 Hz. At last, this work is concluded with a comparative bit error rate study on
the basis of singular parameters at different SNR levels to produce the system performance for uncoded
QPSK modulation.
Opportunistic use of the 2.63.5 ghz band for broadband services in the west a...ijmnct
Wireless technology is increasing rapidly, and the vision of pervasive wireless computing and
communications offers the promise of many societal and individual benefits. While consumer devices such
as cell phones, PDAs and laptops receive a lot of attention, the impact of wireless technology is much
broader, e.g., through sensor networks for safety applications and home automation, smart grid control,
medical wearable and embedded wireless devices, and entertainment systems. One of these wireless
technologies is the Worldwide Interoperability for Microwave Access (WiMAX) technology. The explosion
of wireless applications in some parts of the world has created an ever-increasing demand for more radio
spectrum. This is not the case in the West African Sub-Region, especially Ghana where the 2.6GHz and
3.5GHz broadband access bands offering 190MHz and 140MHz bandwidth is underutilized. In this paper,
we look at usage of deployed 4G-WiMAX network in Ghana and advocate the need for policy to promote
the usage of licensed bands opportunistically by wireless devices and/or networks for application in
security, smart grid control, e-learning, telemedicine, e-governance, home and factory automation
Effective range free localization scheme for wireless sensor networkijmnct
Location aware sensors can be used in many areas such as military and civilian applications. Wireless
Sensor Networks help to identify the accurate location of the event. In this paper a cost effective schema for
localization has been proposed. It uses two beacon nodes to identify the location of unknown nodes. It
also uses flooding and estimating method to accurately identify the location of other nodes. Available area
is divided into zones and beacons are provided for each zone. Beacon nodes are placed in appropriate
locations normally two in a zone to provide location information. Using the two nodes location of unknown
nodes can be calculated accurately.
A scalable and power efficient solution for routing in mobile ad hoc network ...ijmnct
Mobile Ad Hoc Network (MANET) is a very dynamic and infrastructure-less ad hoc network. The actual
network size depends on the application and the protocols developed for the routing for this kind of
networks should be scalable. MANET is a resource limited network and therefore the developed routing
algorithm for packet transmission should be power and bandwidth efficient. These kinds of dynamic
networks should operate with minimal management overhead. The management functionality of the
network increases with number of nodes and reduces the performance of the network. Here, in this paper,
we have designed all identical nodes in the cluster except the cluster head and this criterion reduces the
management burden on the network. Graph theoretic routing algorithm is used to develop route for packet
transmission by using the minimum resources. In this paper, we developed routing algorithm for cluster
based MANET and finds a path from source to destination using minimum cumulative degree path. Our
simulation results show that this routing algorithm provide efficient routing path with the increasing
number of nodes and uses multi-hop connectivity for intra-cluster to utilize minimum power for packet
transmission irrespective of number of nodes in the network.
ANDROID APPLICATION DEVELOPMENT FOR ENVIRONMENT MONITORING USING SMART PHONESijmnct
Mobile devices (in particular smart phones and tablets) can be used to monitor quality of life parameters.
Today mobile devices use embedded sensors such as accelerometers, compasses, GPSs, microphones, and
cameras without considering, for example, the air quality or the pollutants of the environment. This paper
presents the possibility to use the smart phones capabilities to gather data from other phones or sensors.
The environment condition’s parameters such as temperature and humidity should be monitored. This point
can be obtained by using distributed devices in different environments that containing high-resolution
sensors and a wireless transmission apparatus for transferring data to smart phones. The Bluetooth was
chosen as a transmission tool since it is embedded in all smart phones
Routing in Wireless Mesh Networks: Two Soft Computing Based Approachesijmnct
Due to dynamic network conditions, routing is the most critical part in WMNs and needs to be optimised.
The routing strategies developed for WMNs must be efficient to make it an operationally self configurable
network. Thus we need to resort to near shortest path evaluation. This lays down the requirement of some
soft computing approaches such that a near shortest path is available in an affordable computing time. This
paper proposes a Fuzzy Logic based integrated cost measure in terms of delay, throughput and jitter.
Based upon this distance (cost) between two adjacent nodes we evaluate minimal shortest path that updates
routing tables. We apply two recent soft computing approaches namely Big Bang Big Crunch (BB-BC) and
Biogeography Based Optimization (BBO) approaches to enumerate shortest or near short paths. BB-BC
theory is related with the evolution of the universe whereas BBO is inspired by dynamical equilibrium in
the number of species on an island. Both the algorithms have low computational time and high convergence
speed. Simulation results show that the proposed routing algorithms find the optimal shortest path taking
into account three most important parameters of network dynamics. It has been further observed that for
the shortest path problem BB-BC outperforms BBO in terms of speed and percent error between the
evaluated minimal path and the actual shortest path.
REMOTE MEASUREMENT SYSTEM GROUND SHIFT WITH GSMijmnct
Indonesia is a country that a lot of valleys, hills and volcanoes. So every year, Indonesia many natural
disasters landslides. A landslide is the movement of slope-forming materials such as rocks, debris material,
soil, or a mixture of materials, moving down or off the slopes. This research tries to build an early warning
system of landslides using microcontroller ATMEGA8535, using draw wire sensor and telemetry using
Siemens mobile phones. At ground shifted over 4 cm then this system will sound sirens and danger will
contact the village to evacuate its citizens. The results showed that the system can transmit data via GSM
network information to the media in the form of SMS time data (day, month, year, hour, and minute) with
soil shear scale of 1 mm. The system can transmit data in the form of SMS information automatically when
shifting ground beyond the normal values determined shift to one of the numbers that have been registered
as a shipping destination number
Improving energy efficiency in manet’s for healthcare environmentsijmnct
Now a day ad hoc mobile networks (MANETs) have lots of routing protocols, but no one can meet
maximum performance. Some are good in a small network; some are suitable in large networks, and some
give better performance in location or global networks. Today modern and innovative applications for
health care environments based on a wireless network are being developed in the commercial sectors. The
emerging wireless networks are rapidly becoming a fundamental part of every single field of life. Our
proposed DEERP framework gives a better performance as compared to other routing protocol.
Optimized rationalize security and efficient data gathering in wireless senso...ijmnct
Wireless reprogramming during a wireless detector net- work (WSN) is that the method of propagating a
replacement code image or relevant commands to detector nodes. As a WSN is sometimes deployed in
hostile environments, secure reprogramming is and can continue to be a significant concern. Whereas all
existing insecure/secure reprogramming protocols square measure supported the centralized approach,
it\'s necessary to support distributed reprogramming during which multiple licensed network users will at
the same time and directly reprogram detector nodes while not involving the bottom station. Terribly
recently, a novel secure rationalize and distributed reprogramming protocol named SRDRP has been
planned, that is that the initial work of its kind. However, during this paper, we have a tendency to establish
associate inherent style weakness within the increased signature verification of SRDRP associated demonstrate
that it\'s at risk of associate impersonation attack by that an resister will simply impersonate any
licensed user to hold out reprogramming. Later on, we have a tendency to propose a straightforward
modification to mend the known security drawback while not losing any options of SRDRP. Our
experimental results demonstrate that it\'s able to eliminate the planning weakness by adding one-B
redundant information which the execution time of the prompt answer during a 1.6-GHz laptop personal
computer is not any quite one ms. Therefore, our answer is possible and secure for real-world applications.
Moreover, we have a tendency to show that, so as to additional improve the safety and potency of SRDRP;
any higher established identity-based position formula will be directly utilized in SRDRP. Supported
implementation results, we have a tendency to demonstrate potency improvement over the initial SRDRP
Cluster head election using imperialist competitive algorithm (chei) for wire...ijmnct
One of the most important challenges of wireless sensor network is how to prolong its life time. The main
obstacle in these networks is the limited energy of nodes. We can overcome this problem by optimizing the
nodes' power consumption. The clustering mechanismis the one of the representative approachesto reduce
energy consumption, but optimum clustering of wireless sensor network is an NP-Hard problem. This
paper proposes a hybrid algorithm based on Imperialist competitive algorithm to overcome this clustering
problem. The proposed method, acts on one of the clusters in the network to choose the best sensor in
the cluster as a cluster head. To perform this action, the cluster is divided into several sub-clusters,
each of which has a cluster head. These cluster heads using Assimilation
policies, try to attract the regular nodes to themselves, and Using Imperialistic competition,
they compete with each other until one of these cluster heads is selected as the final cluster head. After this
stage, the algorithm work ends. This algorithm will balance the energy consumption in the network and
improve the network lifetime. To prove efficiency of proposed algorithm(CHEI), we simulated the proposed
algorithm compared with two clustering algorithms using the matlab
Improvement in the mobility of mobile ipv6 based mobile networks using revers...ijmnct
Mobile IPv6 based Mobile Networks are becoming increasingly important with the widespread
popularity of wireless internet connectivity. For large networks, an extension, namely, the Hierarchical
Mobile IPv6 is used, which suffers from overloading of the Home Agent as every packet sent has to pass
through it and this produces a significant loss in performance. In this paper, a method to improve the
mobility using the fast handover of Mobile IPv6 (FMIPv6) and the Reverse Routing Header Protocol is
proposed. Simulation results at different packet intervals show that, the proposed scheme is able to
reduce the average delay and achieve an optimum level of throughput.
Multilayer Perceptron Guided Key Generation through Mutation with Recursive R...pijans
In this paper, a multilayer perceptron guided key generation for encryption/decryption (MLPKG) has been
proposed through recursive replacement using mutated character code generation for wireless
communication of data/information. Multilayer perceptron transmitting systems at both ends accept an
identical input vector, generate an output bit and the network are trained based on the output bit which is
used to form a protected variable length secret-key. For each session, different hidden layer of multilayer
neural network is selected randomly and weights or hidden units of this selected hidden layer help to form
a secret session key. The plain text is encrypted using mutated character code table. Intermediate cipher
text is yet again encrypted through recursive replacement technique to from next intermediate encrypted
text which is again encrypted to form the final cipher text through chaining , cascaded xoring of multilayer
perceptron generated session key. If size of the final block of intermediate cipher text is less than the size of
the key then this block is kept unaltered. Receiver will use identical multilayer perceptron generated
session key for performing deciphering process for getting the recursive replacement encrypted cipher text
and then mutated character code table is used for decoding. Parametric tests have been done and results
are compared in terms of Chi-Square test, response time in transmission with some existing classical
techniques, which shows comparable results for the proposed technique.
SYMMETRIC KEY MANAGEMENT SCHEME FOR HIERARCHICAL WIRELESS SENSOR NETWORKSIJNSA Journal
Wireless Sensor Networks (WSNs) are critical component in many applications that used for data collection. Since sensors have limited resource, Wireless Sensor Networks are more vulnerable to attacks than other wireless networks. It is necessary to design a powerful key management scheme for WSNs and take in consideration the limited characteristics of sensors. To achieve security of communicated data in the network and to extend the WSNs lifetime; this paper proposes a new scheme called Symmetric Key Management Scheme (SKMS). SKMS used Symmetric Key Cryptography that depends only on a Hash function and XOR operation for securing homogeneous and heterogeneous hierarchical WSNs. Symmetric Key Cryptography is less computation than Asymmetric Key Cryptography. Simulation results show that the proposed scheme provides security, save the energy of sensors with low computation overhead.
SYMMETRIC KEY MANAGEMENT SCHEME FOR HIERARCHICAL WIRELESS SENSOR NETWORKSIJNSA Journal
Wireless Sensor Networks (WSNs) are critical component in many applications that used for data
collection. Since sensors have limited resource, Wireless Sensor Networks are more vulnerable to
attacks than other wireless networks. It is necessary to design a powerful key management scheme for WSNs
and take in consideration the limited characteristics of sensors. To achieve security of communicated
data in the network and to extend the WSNs lifetime; this paper proposes a new scheme called
Symmetric Key Management Scheme (SKMS). SKMS used Symmetric Key Cryptography that depends
only on a Hash function and XOR operation for securing homogeneous and heterogeneous hierarchical
WSNs. Symmetric Key Cryptography is less computation than Asymmetric Key Cryptography. Simulation
results show that the proposed scheme provides security, save the energy of sensors with low
computation overhead.
ARTIFICIAL NEURAL NETWORK APPROACH TO MODELING OF POLYPROPYLENE REACTORijac123
This paper shows modeling of highly nonlinear polymerization process using the artificial neural network approach for the model predictive purposes. Polymerization occurs in a fluidized bed polypropylene reactor using Ziegler - Natta catalyst and the main objective was modeling of the reactor production rate.
The data set used for an identification of the model is a real process data received from an existing polypropylene plant and the identified model is a nonlinear autoregressive neural network with the exogenous input. Performance of a trained network has been verified using the real process data and the
ability of the production rate prediction is shown in the conclusion.
Biomedical Signals Classification With Transformer Based Model.pptxSandeep Kumar
Seizures are caused by abnormal neuronal activity and are a common chronic brain disease. It is estimated that around 60 million people have epileptic seizures worldwide. Having an epileptic seizure can have serious consequences for the patient. Surface electroencephalogram (EEG) is a non-intrusive method frequently used to detect epileptic brain action. Nevertheless, graphic inspection of the EEG is biased, time-taken, and tedious for the neurologist. This paper proposes an automatic epileptic seizure classification technique using transformer based deep learning. Fast Fourier Transform (FFT) is first used to extract features from EEG data, and features are then used as inputs to a classifier for selection and classification. The suggested technique successfully gave 100% accuracy in the tested EEG reading. The effectiveness of the proposed method could vary across different EEG databases.
A Mixed Binary-Real NSGA II Algorithm Ensuring Both Accuracy and Interpretabi...IJECEIAES
In this work, a Neuro-Fuzzy Controller network, called NFC that implements a Mamdani fuzzy inference system is proposed. This network includes neurons able to perform fundamental fuzzy operations. Connections between neurons are weighted through binary and real weights. Then a mixed binaryreal Non dominated Sorting Genetic Algorithm II (NSGA II) is used to perform both accuracy and interpretability of the NFC by minimizing two objective functions; one objective relates to the number of rules, for compactness, while the second is the mean square error, for accuracy. In order to preserve interpretability of fuzzy rules during the optimization process, some constraints are imposed. The approach is tested on two control examples: a single input single output (SISO) system and a multivariable (MIMO) system.
Implementation of Feed Forward Neural Network for Classification by Education...ijsrd.com
in the last few years, the electronic devices production field has witness a great revolution by having the new birth of the extraordinary FPGA (Field Programmable Gate Array) family platforms. These platforms are the optimum and best choice for the modern digital systems now a day. The parallel structure of a neural network makes it potentially fast for the computation of certain tasks. The same feature makes a neural network well suited for implementation in VLSI technology. In this paper a hardware design of an artificial neural network on Field Programmable Gate Arrays (FPGA) is presented. Digital system architecture is designed to realize a feed forward multilayer neural network. The designed architecture is described using Very High Speed Integrated Circuits Hardware Description Language (VHDL).General Terms-Network.
In this paper a novel intelligent soft computing based cryptographic technique based on synchronization of
two chaotic systems (CSCT) between sender and receiver has been proposed to generate session key using
Pecora and Caroll (PC) method. Chaotic system has some unique features like sensitive to initial
conditions, topologically mixing; and dense periodic orbits. By nature, the Lorenz system is very sensitive
to initial conditions meaning that the error between attacker and receiver is going to grow exponentially if
there is a very slight difference between their initial conditions. All these features make chaotic system as
good alternatives for session key generation. In the proposed CSCT few parameters ( , b , r , x1 ,y2 and z2 )
are being exchanged between sender and receiver. Some of the parameter which takes major roles to form
the session key does not get transmitted via public channel, sender keeps these parameters secret. This way
of handling parameter passing mechanism prevents any kind of attacks during exchange of parameters like
sniffing, spoofing or phishing.
Web Spam Classification Using Supervised Artificial Neural Network Algorithmsaciijournal
Due to the rapid growth in technology employed by the spammers, there is a need of classifiers that are
more efficient, generic and highly adaptive. Neural Network based technologies have high ability of
adaption as well as generalization. As per our knowledge, very little work has been done in this field using
neural network. We present this paper to fill this gap. This paper evaluates performance of three supervised
learning algorithms of artificial neural network by creating classifiers for the complex problem of latest
web spam pattern classification. These algorithms are Conjugate Gradient algorithm, Resilient Backpropagation learning, and Levenberg-Marquardt algorithm.
Differential Protection of Generator by Using Neural Network, Fuzzy Neural an...Waqas Tariq
This paper describes Artificial Intelligences techniques for detecting internal faults in a generator. Three techniques uses Neural Net (NN), Fuzzy Neural (FNN) and Fuzzy Neural Petri Net (FNPN) to implements differential protection of generator. MATLAB toolbox has been used for simulations and generation of fault data to training the programs for different faults cases and to implement the relay. Result of different cases studies are presented and compares among three implements techniques.
Soft computing based cryptographic technique using kohonen's selforganizing m...ijfcstjournal
In this paper a novel soft computing based cryptographic technique based on synchronization of two
Kohonen's Self-Organizing Feature Map (KSOFM) between sender and receiver has been proposed. In this
proposed technique KSOFM based synchronization is performed for tuning both sender and receiver. After
the completion of the tuning identical session key get generates at the both sender and receiver end with the
help of synchronized KSOFM. This synchronized network can be used for transmitting message using any
light weight encryption/decryption algorithm with the help of identical session key of the synchronized
network. Exhaustive parametric tests are done and results are compared with some existing classical
techniques, which show comparable results for the proposed system.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
Neural Cryptography for Secret Key ExchangeIJMTST Journal
The goal of any cryptography system is the exchange of information among the intended user without any leakage of information to other who may have unauthorized access to it. A common secret key could be created over a public channel accessible to any opponent. Neural networks can be used to generate common secret key. In case of neural cryptography, both the communicating networks receive an identical input vector, generate an output bit and are trained based on the output bit. The two networks and their weights vectors exhibit a new phenomenon, where the networks synchronize to a state with identical time-dependent weights. The generated secret key over a public channel is used for encryption and decryption of the message or information send over the channel.
Similar to Secured transmission through multi layer perceptron in wireless communication (stmlp) (20)
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AIVladimir Iglovikov, Ph.D.
Presented by Vladimir Iglovikov:
- https://www.linkedin.com/in/iglovikov/
- https://x.com/viglovikov
- https://www.instagram.com/ternaus/
This presentation delves into the journey of Albumentations.ai, a highly successful open-source library for data augmentation.
Created out of a necessity for superior performance in Kaggle competitions, Albumentations has grown to become a widely used tool among data scientists and machine learning practitioners.
This case study covers various aspects, including:
People: The contributors and community that have supported Albumentations.
Metrics: The success indicators such as downloads, daily active users, GitHub stars, and financial contributions.
Challenges: The hurdles in monetizing open-source projects and measuring user engagement.
Development Practices: Best practices for creating, maintaining, and scaling open-source libraries, including code hygiene, CI/CD, and fast iteration.
Community Building: Strategies for making adoption easy, iterating quickly, and fostering a vibrant, engaged community.
Marketing: Both online and offline marketing tactics, focusing on real, impactful interactions and collaborations.
Mental Health: Maintaining balance and not feeling pressured by user demands.
Key insights include the importance of automation, making the adoption process seamless, and leveraging offline interactions for marketing. The presentation also emphasizes the need for continuous small improvements and building a friendly, inclusive community that contributes to the project's growth.
Vladimir Iglovikov brings his extensive experience as a Kaggle Grandmaster, ex-Staff ML Engineer at Lyft, sharing valuable lessons and practical advice for anyone looking to enhance the adoption of their open-source projects.
Explore more about Albumentations and join the community at:
GitHub: https://github.com/albumentations-team/albumentations
Website: https://albumentations.ai/
LinkedIn: https://www.linkedin.com/company/100504475
Twitter: https://x.com/albumentations
In his public lecture, Christian Timmerer provides insights into the fascinating history of video streaming, starting from its humble beginnings before YouTube to the groundbreaking technologies that now dominate platforms like Netflix and ORF ON. Timmerer also presents provocative contributions of his own that have significantly influenced the industry. He concludes by looking at future challenges and invites the audience to join in a discussion.
In the rapidly evolving landscape of technologies, XML continues to play a vital role in structuring, storing, and transporting data across diverse systems. The recent advancements in artificial intelligence (AI) present new methodologies for enhancing XML development workflows, introducing efficiency, automation, and intelligent capabilities. This presentation will outline the scope and perspective of utilizing AI in XML development. The potential benefits and the possible pitfalls will be highlighted, providing a balanced view of the subject.
We will explore the capabilities of AI in understanding XML markup languages and autonomously creating structured XML content. Additionally, we will examine the capacity of AI to enrich plain text with appropriate XML markup. Practical examples and methodological guidelines will be provided to elucidate how AI can be effectively prompted to interpret and generate accurate XML markup.
Further emphasis will be placed on the role of AI in developing XSLT, or schemas such as XSD and Schematron. We will address the techniques and strategies adopted to create prompts for generating code, explaining code, or refactoring the code, and the results achieved.
The discussion will extend to how AI can be used to transform XML content. In particular, the focus will be on the use of AI XPath extension functions in XSLT, Schematron, Schematron Quick Fixes, or for XML content refactoring.
The presentation aims to deliver a comprehensive overview of AI usage in XML development, providing attendees with the necessary knowledge to make informed decisions. Whether you’re at the early stages of adopting AI or considering integrating it in advanced XML development, this presentation will cover all levels of expertise.
By highlighting the potential advantages and challenges of integrating AI with XML development tools and languages, the presentation seeks to inspire thoughtful conversation around the future of XML development. We’ll not only delve into the technical aspects of AI-powered XML development but also discuss practical implications and possible future directions.
Observability Concepts EVERY Developer Should Know -- DeveloperWeek Europe.pdfPaige Cruz
Monitoring and observability aren’t traditionally found in software curriculums and many of us cobble this knowledge together from whatever vendor or ecosystem we were first introduced to and whatever is a part of your current company’s observability stack.
While the dev and ops silo continues to crumble….many organizations still relegate monitoring & observability as the purview of ops, infra and SRE teams. This is a mistake - achieving a highly observable system requires collaboration up and down the stack.
I, a former op, would like to extend an invitation to all application developers to join the observability party will share these foundational concepts to build on:
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!SOFTTECHHUB
As the digital landscape continually evolves, operating systems play a critical role in shaping user experiences and productivity. The launch of Nitrux Linux 3.5.0 marks a significant milestone, offering a robust alternative to traditional systems such as Windows 11. This article delves into the essence of Nitrux Linux 3.5.0, exploring its unique features, advantages, and how it stands as a compelling choice for both casual users and tech enthusiasts.
Maruthi Prithivirajan, Head of ASEAN & IN Solution Architecture, Neo4j
Get an inside look at the latest Neo4j innovations that enable relationship-driven intelligence at scale. Learn more about the newest cloud integrations and product enhancements that make Neo4j an essential choice for developers building apps with interconnected data and generative AI.
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...Neo4j
Leonard Jayamohan, Partner & Generative AI Lead, Deloitte
This keynote will reveal how Deloitte leverages Neo4j’s graph power for groundbreaking digital twin solutions, achieving a staggering 100x performance boost. Discover the essential role knowledge graphs play in successful generative AI implementations. Plus, get an exclusive look at an innovative Neo4j + Generative AI solution Deloitte is developing in-house.
UiPath Test Automation using UiPath Test Suite series, part 5DianaGray10
Welcome to UiPath Test Automation using UiPath Test Suite series part 5. In this session, we will cover CI/CD with devops.
Topics covered:
CI/CD with in UiPath
End-to-end overview of CI/CD pipeline with Azure devops
Speaker:
Lyndsey Byblow, Test Suite Sales Engineer @ UiPath, Inc.
zkStudyClub - Reef: Fast Succinct Non-Interactive Zero-Knowledge Regex ProofsAlex Pruden
This paper presents Reef, a system for generating publicly verifiable succinct non-interactive zero-knowledge proofs that a committed document matches or does not match a regular expression. We describe applications such as proving the strength of passwords, the provenance of email despite redactions, the validity of oblivious DNS queries, and the existence of mutations in DNA. Reef supports the Perl Compatible Regular Expression syntax, including wildcards, alternation, ranges, capture groups, Kleene star, negations, and lookarounds. Reef introduces a new type of automata, Skipping Alternating Finite Automata (SAFA), that skips irrelevant parts of a document when producing proofs without undermining soundness, and instantiates SAFA with a lookup argument. Our experimental evaluation confirms that Reef can generate proofs for documents with 32M characters; the proofs are small and cheap to verify (under a second).
Paper: https://eprint.iacr.org/2023/1886
Threats to mobile devices are more prevalent and increasing in scope and complexity. Users of mobile devices desire to take full advantage of the features
available on those devices, but many of the features provide convenience and capability but sacrifice security. This best practices guide outlines steps the users can take to better protect personal devices and information.
DevOps and Testing slides at DASA ConnectKari Kakkonen
My and Rik Marselis slides at 30.5.2024 DASA Connect conference. We discuss about what is testing, then what is agile testing and finally what is Testing in DevOps. Finally we had lovely workshop with the participants trying to find out different ways to think about quality and testing in different parts of the DevOps infinity loop.
The Art of the Pitch: WordPress Relationships and SalesLaura Byrne
Clients don’t know what they don’t know. What web solutions are right for them? How does WordPress come into the picture? How do you make sure you understand scope and timeline? What do you do if sometime changes?
All these questions and more will be explored as we talk about matching clients’ needs with what your agency offers without pulling teeth or pulling your hair out. Practical tips, and strategies for successful relationship building that leads to closing the deal.
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Albert Hoitingh
In this session I delve into the encryption technology used in Microsoft 365 and Microsoft Purview. Including the concepts of Customer Key and Double Key Encryption.
Unlocking Productivity: Leveraging the Potential of Copilot in Microsoft 365, a presentation by Christoforos Vlachos, Senior Solutions Manager – Modern Workplace, Uni Systems
Secured transmission through multi layer perceptron in wireless communication (stmlp)
1. International Journal of Mobile Network Communications & Telematics ( IJMNCT) Vol. 4, No.4,August 2014
SECURED TRANSMISSION THROUGH MULTI LAYER
PERCEPTRON IN WIRELESS COMMUNICATION
(STMLP)
Arindam Sarkar1 and J. K. Mandal2
1Department of Computer Science & Engineering, University of Kalyani, W.B, India
2Department of Computer Science & Engineering, University of Kalyani, W.B, India
ABSTRACT
In this paper, a multilayer perceptron guided encryption/decryption (STMLP) in wireless communication
has been proposed for exchange of data/information. Multilayer perceptron transmitting systems at both
ends generate an identical output bit and the network are trained based on the output which is used to
synchronize the network at both ends and thus forms a secret-key at end of synchronizations of the
networks. Weights or hidden units of the hidden layer help to form a secret session key. The plain text is
encrypted through chaining , cascaded xoring of multilayer perceptron generated session key. If size of the
final block of plain text is less than the size of the key then this block is kept unaltered. Receiver will use
identical multilayer perceptron generated session key for performing deciphering process for getting the
plain text. Parametric tests have been done and results are compared in terms of Chi-Square test, response
time in transmission with some existing classical techniques, which shows comparable results for the
proposed technique. Variation numbers of input vectors and hidden layers will increase the confusion
/diffusion of the schemeand hence increase the security. As a result variable energy based techniques may
be achieved which may be applicable devices/interface of the heterogeneous sizes of the network/device.
KEYWORDS
Multilayer Perceptron, Session Key, Wireless Communication.
1. INTRODUCTION
In recent times wide ranges of techniques are developed to protect data and information from
eavesdroppers [1, 2, 3, 4, 5, 6, 7, 8, 9]. Algorithms have their virtue and shortcomings. For
Example in DES, AES algorithms [1] the cipher block length is nonflexible. In NSKTE [4],
NWSKE [5], AGKNE [6], ANNRPMS [7] and ANNRBLC [8] technique uses two neural
network one for sender and another for receiver having one hidden layer for producing
synchronized weight vector for key generation. Attacker can get an idea about sender and
receiver’s neural machine as session architecture of neural machine is static. In NNSKECC
algorithm [9] any intermediate blocks throughout its cycle taken as the encrypted block and this
number of iterations acts as secret key. Here, if n number of iterations are needed for cycle
formation and if intermediate block is chosen as an encrypted block after n/2th iteration then
exactly same number of iterations i.e. n/2 are needed for decode the block which makes easier
the attackers life. This paper proposed a multilayer perceptron guided encryption technique in
wireless communication to overcome the problem.
The organization of this paper is as follows. Section 2 of the paper deals with the problem domain
and methodology. Proposed Multilayer Perceptron based key generation has been discussed in
DOI : 10.5121/ijmnct.2014.4401 1
2. International Journal of Mobile Network Communications & Telematics ( IJMNCT) Vol. 4, No.4,August 2014
section 3. Triangularization encryption technique is given in section 4. Triangularization
decryption has been presented in section 5. Section 6 presents energy computation technique.
Complexity analysis of the technique is given in section 7. Experimental results are described in
section 8. Analysis of the results presented in section 9. Analysis regarding various aspects of the
technique has been presented in section 10. Conclusions and future scope are drawn in section 11
and that of references at end.
2
2. PROBLEM DOMAIN AND METHODOLOGY
In security based communication the main problem is distribution of key between sender and
receiver. As, during exchange of key over public channel intruders can intercept the key as a
middleman. The problem has been addressed and a technique has been proposed addressing the
issue. These are presented in section 2.1 and 2.2 respectively.
2.1. Man-In-The-Middle Attack
Intruders intercepting in the middle between sender and receiver try to capture all the information
transmitting from both. Diffie-Hellman key exchange technique [1] suffers from this type of
problems. Intruders can act as sender/ receiver simultaneously and try to steal secret session key
at the time of exchanging key via public channel.
2.2. Methodology
Well known problem of man in the middle attack has been addressed in STMLP where secret
session key is not exchanged over public insecure channel. At end of synchronization both
parties’ generates identical weight vectors and activated hidden layer outputs for both the parties
become identical. This identical output of hidden layer for both parties are used as one time secret
session key for secured data exchange.
The basic idea here is to design such a program with effective GUI which helps people to
understand the underlying calculations. In this case this would be the Tree Parity Machine and the
various encryption and decryption techniques. First we need to figure out what are main functions
of our system. Since we are going to work on various Neural network structures we need a menu
to choose from. Again after that, two different Neural network need mutual synchronization and
associated statistical data like type of network and total time required to synchronize mutually.
Then at the end we need a menu to choose various encryption and decryption techniques and
statistical modules to compute probable power consumption by the network. So we need various
menus / forms to cater our need of various functions within their scope. So the schematic view
looks like the figure 1.
3. International Journal of Mobile Network Communications & Telematics ( IJMNCT) Vol. 4, No.4,August 2014
3
Figure 1.
Selection of Type
of Neural Network
Encryption
/Decryptio
Analysis
Module
K,N,L Tuning
Time
Data Base
Statistical Data
Figure 1. Schematic diagram
3.MULTILAYER PERCEPTRON BASED SESSION KEY
GENERATION
A multilayer perceptron synaptic simulated weight based undisclosed key generation is carried
out between recipient and sender. Figure 2 shows multilayer perceptron based synaptic simulation
system. Same single hidden layer among multiple hidden layers for a particular session. All other
hidden layers goes in deactivated mode with the incoming input. The key generation technique
with analysis using random number of nodes (neurons) along with the corresponding algorithm is
discussed in the subsections 3.1 to 3.5.
Figure 2. A multilayer perceptron with 3 hidden Layers
Multilayer perceptron in each session acts as a single layer network with dynamically chosen one
activated hidden layer and K no. of hidden neurons, N no. of input neurons having binary input
vector, Î{−1,+1} ij x , discrete weights, are generated from input to output, are lies between -L and
+L, w { L L L} ij Î − ,− +1,...,+ .Where i = 1,…,K denotes the ith hidden unit of the perceptron and j = 1,…,N
the elements of the vector and one output neuron. Output of the hidden units is calculated by the
4. International Journal of Mobile Network Communications & Telematics ( IJMNCT) Vol. 4, No.4,August 2014
weighted sum over the current input values . So, the state of the each hidden neurons is expressed
using (eq.1)
4
= = (1)
h w x
i i i i j,
i j
N
j
N
wx
N
1
,
1 1
=
Output of the ith hidden unit is defined as
sgn( ) i i s = h
(2)
In case of i h = 0, i s = -1 to produce a binary output. Hence, i s = +1, if the weighted sum over its
inputs is positive, or else it is inactive, i s = -1. The output of a perceptron is the product of the
hidden units expressed in (eq. 2).
t s (3)
Õ=
=
K
i
i
1
3.1 Simulation
Input:-Random weights, input vectors for both multilayer perceptrons.
Output:-Secret key through synchronization of input and output neurons as vectors. Method:-
Random vectors generated, fed into the networks. Vectors are updated only when output of
machines produce identical output . The process continue till both machines are fully
synchronized.
Step 1. Initialize of random weight values of synaptic links between input layer and
randomly selected activated hidden layer.
Where, { } L L L wi
j Î− ,− +1,...+, (4)
Step 2. Repeat step 3 to 6 until the full synchronization is achieved, using
Hebbian-learning rules.
( ( ) ( A B ))
i j i j i w g w x i j = + tQs t Qt t +
, , , (5)
Step 3. Generate random input vector X. Inputs are generated by a third party or one of the
communicating parties.
Step 4. Compute the values of the activated hidden neurons of activated hidden layer using
(eq. 6)
= = (6)
h w x
i i i i j,
i j
N
j
N
wx
N
1
,
1 1
=
Step 5. Compute the value of the output neuron using
t s (7)
Õ=
=
K
i
i
1
Compare the output values of both multilayer perceptron by exchanging the system
outputs.
if Output (A) Output (B), Go to step 3
else if Output (A) = Output (B) then one of the suitable learning rule is applied only
the hidden units are trained which have an output bit identical to the
common output.
Update the weights only if the final output values of the perceptron are equivalent. When
synchronization is finally achieved, the synaptic weights are identical for both the system.
5. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
5
i j i w g w f x i j , , , , , = + s t t +
Hebbian learning
anti-Hebbian learning
Random walk learning
3.2 Multilayer Perceptron Learning
At the beginning of the synchronization process multilayer perceptron of A and B start with
uncorrelated weight vectors A B
i w / . For each time step K, public input vectors are generated
randomly and the corresponding output bits t
A/Bare calculated. Afterwards A and B communicate
their output bits to each other. If they disagree, t
A
t
B, the weights are not changed. Otherwise
learning rules suitable for synchronization is applied. In the case of the Hebbian learning rule [10]
both neural networks learn from each other.
( ( ) ( )) A B
i j i j i wi j g w x = + tQst Qt t +
, , , (8)
The learning rules used for synchronizing multilayer perceptron share a common structure. That
is why they can be described by a single (eq. 4)
( ( A B
) ) i j
(9)
with a function ( A B )
i f s ,t ,t , which can take the values -1, 0, or +1. In the case of bidirectional
interaction it is given by
s
( ) ( ) ( )
, , =Q Q −
s
1
s t t st t t A B A A B
i f
(10)
The common part ( A ) ( A B ) Qst Qt t of ( A B )
i f s ,t ,t controls, when the weight vector of a hidden
unit is adjusted. Because it is responsible for the occurrence of attractive and repulsive steps [6].
3.3 Weight Distribution within Multilayer Perceptron
In case of the Hebbian rule (eq. 8), A's and B's multilayer perceptron learn their own output.
Therefore the direction in which the weight i j w , moves is determined by the product i i j x , s . As
the output i s is a function of all input values, i j x , and i s are correlated random variables. Thus
the probabilities to observe i i j x , s = +1 or i i j x , s = -1 are not equal, but depend on the value of the
corresponding weight i j w , [11, 13, 14, 15, 16].
1
w
( )
P s x erf (11)
6.
7.
8. i j
−
= = +
2
,
,
, 1
2
1
i i j
i i j
NQ w
According to this equation, sgn( ) i i, j i, j s x = w occurs more often than the opposite,
sgn( ) i i , j i , j s x = − w . Consequently, the Hebbian learning rule (eq. 8) pushes the weights
towards the boundaries at -L and +L. In order to quantify this effect the stationary probability
9. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
distribution of the weights for t ® ¥ is calculated for the transition probabilities. This leads to
[11].
6
( )
−
( )
Õ=
15. − −
+
= =
w
m
i
i
i j
NQ m
erf
NQ m
erf
P w w P
1
2
2
, 0
1
1
1
1
(12)
Here the normalization constant 0 r is given by
−
( )
1
1
2
2
0
1
1
1
1
−
L
= − =
27. m
− −
+
= Õ
w L
w
m
m
i
i
NQ m
erf
NQ m
erf
P
(13)
In the limit N ® ¥ the argument of the error functions vanishes, so that the weights stay
uniformly distributed. In this case the initial length of the weight vectors is not changed by the
process of synchronization.
( 1
)
3
( 0)
+
= =
L L
Q t i (14)
But if N is finite, the probability distribution itself depends on the order parameter i Q Therefore its
expectation value is given by the solution of the following equation:
L
( )
=−
= =
i i j Q w P w w ,
w L
2
(15)
3.4 Order Parameters
In order to describe the correlations between two multilayer perceptron caused by the
synchronization process, one can look at the probability distribution of the weight values in each
hidden unit. It is given by (2L + 1) variables.
P P(w a w b) B
i
= A
= Ù = a , b i , j
i , j
(16)
which are defined as the probability to find a weight with w a A
i j = , in A's multilayer perceptron and
w b B
i j = , in B's multilayer perceptron. In both cases, simulation and iterative calculation, the
standard order parameters, which are also used for the analysis of online learning, can be
calculated as functions of i
a b P ,
[12].
= − = −
1 2
A
= w w =
a P
i L
Q ,
a L
L
b L
i
a b
A
i
A
i
N
(17)
2 1
= − = −
B
= w w =
b P
i L
Q ,
a L
L
b L
i
a b
B
i
B
i
N
(18)
= − = −
1
AB
= w w =
abP
i L
R ,
a L
L
b L
i
a b
B
i
A
i
N
(19)
28. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
7
The level of synchronization is given by the normalized overlap between two corresponding
hidden units as
r = = (20)
B
i
AB
i
R
A
i
B
i
B
i
w w
A
i
A
i
B
i
A
AB i
i
Q Q
w w w w
3.5 Secret Session Key
At end of full weight synchronization process, weight vectors between input layer and activated
hidden layer of both multilayer perceptron systems become identical. Activated hidden layer’s
output of source multilayer perceptron is used to construct the secret session key. This session key
is not get transmitted over public channel because receiver multilayer perceptron has same
identical activated hidden layer’s output. Compute the values of the each hidden unit by
29.
30. s sgn ( )
N
=
i ij ij w x
=
j
1
−
=
0
1
1
sgn x
if
if
if
0,
0,
0.
=
x
x
x
(21)
For example consider 8 hidden units of activated hidden layer having absolute value (1, 0, 0, 1, 0,
1, 0, 1) becomes an 8 bit block. This 10010101 become a secret session key for a particular
session and cascaded xored with recursive replacement encrypted text. Now final session key
based encrypted text is transmitted to the receiver end. Receiver has the identical session key i.e.
the output of the hidden units of activated hidden layer of receiver. This session key used to get
the recursive replacement encrypted text from the final cipher text. In the next session both the
machines started tuning again to produce another session key.
Identical weight vector derived from synaptic link between input and activated hidden layer of
both multilayer perceptron can also becomes secret session key for a particular session after full
weight synchronization is achieved.
4. ENCRYPTION
For encryption a triangular based technique has been described. During plain text encryption, in
the first phase consider a block S = s0
0 s0
1 s0
2 s0
3 s0
4 s0
5 … s0
n-2 s0
n-1 of size n bits, where s0
i = 0 or
1 for 0 = i = (n-1). Now, starting from MSB (s0
0) and the next-to-MSB (s0
1), bits are pair-wise
XORed, so that the 1st intermediate sub-stream S1 = s1
0 s1
1 s1
2 s1
3 s1
4 s1
5 … s1
n-2 is generated
consisting of (n-1) bits, where s1
j = s0
j Å s0
j+1 for 0 = j = n-2, Å stands for the exclusive OR
operation. This 1st intermediate sub-stream S1 is also then pair-wise XORed to generate S2 = s2
0
s2
1 s2
2 s2
3 s2
4 s2
5 … s2
n-3, which is the 2nd intermediate sub-stream of length (n-2). This process
continues (n-1) times to ultimately generate Sn-1 = sn-1
0, which is a single bit only. Thus the size of
the 1st intermediate sub-stream is one bit less than the source sub-stream; the size of each of the
intermediate sub-streams starting from the 2nd one is one bit less than that of the sub- stream
wherefrom it was generated; and finally the size of the final sub-stream in the process is one bit
less than the final intermediate sub-stream. Table 1 and figure 3 show the process.
31. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
8
Table 1
Options for choosing Target Block from Triangle
Option
Serial No.
Target Block Method of Formation
001
s0
0 s1
0 s2
0 s3
0 s4
0 s5
0 …
sn-2
0 sn-1
0
Taking all the MSBs starting
from the source block till the
last block generated
010
sn-1
0 sn-2
0 sn-3
0 sn-4
0 sn-5
0
… s1
0 s0
0
Taking all the MSBs starting
from the last block generated
till the source block
011
s0
n-1 s1
n-2 s2
n-3 s3
n-4 s4
n-5
… sn-2
1 sn-1
0
Taking all the LSBs starting
from the source block till the
last block generated
100
sn-1
0 sn-2
1 sn-3
2 sn-4
3 sn-5
4
… s1
n-2 s0
n-1
Taking all the LSBs starting
from the last block generated
till the source block
Option Serial No. 010 Option Serial No. 100
Option Serial No. 001 Option Serial No. 011
Figure 3. Options diagram for choosing Target Block from Triangle
Table 1describes different options for choosing target block from triangle. This option is
generated by modulo 4 division of the value of output neuron then adding 1. Then take the binary
version of the decimal no. Each block size is represented by 5 bits and 3 bits are used to denoting
the option no. for that block. So, total 8 bits are used to describe a single block length and option
chosen. For multiple blocks several 8bits are attached together preceded by first 8 bits (28 = 256
blocks can be formed in one session) to describe total no. of block to forms intermediate sub key.
Maximum length of this sub key will be (256 blocks X 8 bits per block) 2048 bits. This sub key is
padded in the front of the encrypted text. Then the multilayer perceptron generated synchronized
one time session key is repeatedly xored with the intermediate traingularized cipher text by
considering same key traingularized cipher text length. This mechanism is performed until all
the blocks get exhausted.
5. DECRYPTION
During decryption, the receiver’s multilayer perceptron generated synchronized one time session
key is xored with the cipher text. The technique of performing xoring is same that was in
encryption process. Finally from the outcomes intermediate encrypted block (E) and sub key
block is extracted and now key is use to decipher the E to get the source stream. To ease the
explanation of decryption technique, let us consider, e0
i-1 = si-1
n-i for 1 = i = n, so that the
encrypted block becomes E = e0
0 e0
1 e0
2 e0
3 e0
4 … e0
n-2 e0
n-1. After the formation of the triangle, for
the purpose of decryption, the block en-1
0 en-2
0 en-3
0 en-4
0 en-5
0 … e1
0 e0
0, i.e., the block constructed
by taking all the MSBs of the blocks starting from the finally generated single-bit block En-1 to E,
are to be taken together and it is to be considered as the decrypted block. Figure 4 show the
32. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
triangle generated and hence the decrypted block obtained. Here the intermediate blocks are
referred to as E1, E2, …, En-2 and the final block generated as En-1.
9
Option Serial No. 010 Option Serial No. 100
Option Serial No. 001 Option Serial No. 011
Figure 4. Generation of Source Block from Target
6. ENERGY VARIABILITY
The proposed schemes have a good potential of energy variability which can be incorporated and
may be adopted on the fly during transmission. We know that energy required for a blue tooth
communication is less than that of a WiFi communication. This variability of energy can be
incorporated into the encryption system through incorporating variable number of hidden layers
and input neurons. Table 2 shows the proposed network sizes for various types of wireless
networks.
Table 2. Network type vs size of neural network
Type of Network
Energy
Availability
Network Size and parameters
Wireless Personal Area
Networks (Bluetooth,
Infrared)
Very Low
No. of input layer neurons= 5 to 10
No. of Hidden layer neurons= 4 to 8
No. of Hidden layer = 1 to 3
Synaptic Depth (L)= +5 to -5
Wireless LAN
Low
No. of input layer neurons= 10 to 25
No. of Hidden layer neurons= 8 to 20
No. of Hidden layer = 3 to 4
Synaptic Depth (L)= +10 to -10
Wireless mesh network Medium
No. of input layer neurons= 25 to 40
No. of Hidden layer neurons= 20 to 35
No. of Hidden layer = 4 to 5
Synaptic Depth (L)= +15 to -15
Wireless MAN Moderate
No. of input layer neurons= 40 to 50
No. of Hidden layer neurons= 35 to 45
No. of Hidden layer = 5 to 6
Synaptic Depth (L)= +20 to -20
Wireless WAN Relatively High
No. of input layer neurons= 50 to 55
No. of Hidden layer neurons= 45 to 50
No. of Hidden layer = 5 to 6
Synaptic Depth (L)= +25 to -25
Cellular network High
No. of input layer neurons= 55 to 100
No. of Hidden layer neurons= 50 to 70
No. of Hidden layer = 5 to 6
Synaptic Depth (L)= +30 to -30
33. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
10
7. COMPLEXITY ANALYSIS
The complexity of the technique will be O(L), can be computed using following three steps.
Step 1. To generate a MLP guided key of length N needs O(N) Computational steps. The
average synchronization time is almost independent of the size N of the networks, at
least up to N=1000.Asymptotically one expects an increase like O (log N).
Step 2. Complexity of the encryption technique is O(L).
Step 2. 1. Triangularization encryption process takes O(L).
Step 2. 2. MLP based encryption technique takes O(L) amount of time.
Step 3. Complexity of the decryption technique is O(L).
Step 3. 1. In MLP based decryption technique, complexity to convert final cipher
text into Tra cipher text T takes O(L).
Step 3. 2. Transformation of cipher text T into the corresponding stream of bits S
= s0 s1 s2 s3 s4…sL-1, which is the source block takes O(L) as this step also takes
constant amount of time for merging s0 s1 s2 s3 s4…sL-1.
So, overall time complexity of the entire technique is O(L).
8. RESULTS
In this section the results of implementation of the proposed STMLP technique has been
presented in terms of encryption decryption time, Chi-Square test, source file size vs. encryption
time along with source file size vs. encrypted file size. The results are also compared with
existing RSA [1] technique, existing ANNRBLC[8] and NNSKECC[9].
Table 3. Encryption / decryption time vs. File size
Encryption Time Decryption Time
Source
Size(bytes)
STMLP NNSKECC[9]
Encrypted
Size(bytes)
STMLP NNSKECC[9]
18432 6. 42 7.85 18432 6.99 7.81
23044 9. 23 10.32 23040 9.27 9.92
35425 14. 62 15.21 35425 14. 47 14.93
36242 14. 72 15.34 36242 15. 19 15.24
59398 25. 11 25.49 59398 24. 34 24.95
Table 3 shows encryption and decryption time with respect to the source and encrypted size
respectively. It is also observed the alternation of the size on encryption.
In figure 5 stream size is represented along X axis and encryption/decryption time is represented
along Y-axis. This graph is not linear, because of different time requirement for finding
appropriate MLP key. It is observed that the decryption time is almost linear, because there is no
MLP key generation process during decryption.
34. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
Chi-Square value
11
Encryption decryption time
9.23
Encryption Decryption
18432 23044 35425 36242 59398
Source size
14.62 14.72
9.27
14.47 15.19
Figure 5 Encryption decryption time against stream size
6.42
25.11
6.99
24.34
30
25
20
15
10
5
0
Table 4 shows Chi-Square value for different source stream size after applying different
encryption algorithms. It is seen that the Chi-Square value of STMLP is better compared to the
algorithm ANNRBLC [8] and comparable to the Chi-Square value of the RSA algorithm.
Table 4. Source size vs. Chi-Square value
Stream Size
(bytes)
Chi-Square value
(TDES) [1]
Chi-Square value in
(STMLP)
1500 1228.5803 2856.2673 2471.0724 5623.14
2500 2948.2285 6582.7259 5645.3462 22638.99
3000 3679.0432 7125.2364 6757.8211 12800.355
3250 4228.2119 7091.1931 6994.6198 15097.77
3500 4242.9165 12731.7231 10572.4673 15284.728
Figure 6 shows graphical representation of table 4.
Chi-Square value
(ANNRBLC) [8]
Figure 6. Chi-Square value against stream size
(RSA) [1]
Table 6 shows total number of iteration needed and number of data being transferred for MLP
key generation process with different numbers of input(N) and activated hidden(H) neurons and
varying synaptic depth(L).
35. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
Table 5. Data Exchanged
. and No. of Iterations For Different Parameters Value
Synaptic
Weight (L)
Total No. of
Iterations
Following figure 7. Shows the snapshot of MLP key simulation process. This snapshot shows the
tunning process of two multilayer perceptron with 4 hidden neurons, 4 input neurons and 6 as a
weight value with hebbian learning rule.
Figure 7.
Figure 8 shows the encryption and decryption time of a .txt file. File size taking as a bytes and
encryption/ decryption time represnts as a nanosecond.
Figure 8.
No. of
Input
Neurons(N)
No. of Activated
Hidden Neurons(K)
5 15
30 4
25 5
20 10
8 15
MLP Key Simulation Snapshot
Snapshot of Encryption and decryption time
Data
Exchanged
(Kb)
3 624 48
4 848 102
3 241 30
3 1390 276
4 2390 289
12
36. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
Figure 9 shows the memory heap representation of the key generation technique. The violet color
area represents the memory that has alreay been allocated. Another color represents the total
available memory.
13
Figure 9. Memory Map for whole application
Figure 10 shows the memory allocation gantt chart during key generation process.
Figure 10. Memory Gantt Chart
9. ANALYSIS
From results obtained it is clear that the technique will achieve optimal performances. Encryption
time and decryption time varies almost linearly with respect to the block size. For the algorithm
presented, Chi-Square value is very high compared to some existing algorithms. A user input key
has to transmit over the public channel all the way to the receiver for performing the decryption
procedure. So there is a likelihood of attack at the time of key exchange. To defeat this insecure
secret key generation technique a neural network based secret key generation technique has been
devised. The security issue of existing algorithm can be improved by using MLP secret session
key generation technique. In this case, the two partners A and B do not have to share a common
secret but use their indistinguishable weights or output of activated hidden layer as a secret key
needed for encryption. The fundamental conception of MLP based key exchange protocol focuses
mostly on two key attributes of MLP. Firstly, two nodes coupled over a public channel will
synchronize even though each individual network exhibits disorganized behaviour. Secondly, an
outside network, even if identical to the two communicating networks, will find it exceptionally
difficult to synchronize with those parties, those parties are communicating over a public
network. An attacker E who knows all the particulars of the algorithm and records through this
channel finds it thorny to synchronize with the parties, and hence to calculate the common secret
key. Synchronization by mutual learning (A and B) is much quicker than learning by listening (E)
[10]. For usual cryptographic systems, we can improve the safety of the protocol by increasing of
37. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
the key length. In the case of MLP, we improved it by increasing the synaptic depth L of the
neural networks. For a brute force attack using K hidden neurons, K*N input neurons and
boundary of weights L, gives (2L+1)KN possibilities. For example, the configuration K = 3, L =
3 and N = 100 gives us 3*10253 key possibilities, making the attack unfeasible with today’s
computer power. E could start from all of the (2L+1)3N initial weight vectors and calculate the
ones which are consistent with the input/output sequence. It has been shown, that all of these
initial states move towards the same final weight vector, the key is unique. This is not true for
simple perceptron the most unbeaten cryptanalysis has two supplementary ingredients first; a
group of attacker is used. Second, E makes extra training steps when A and B are quiet [10]-[12].
So increasing synaptic depth L of the MLP we can make our MLP safe.
14
10. SECURITY ISSUE
The main difference between the partners and the attacker in MLP is that A and B are able to
influence each other by communicating their output bits A t B t while E can only listen to these
messages. Of course, A and B use their advantage to select suitable input vectors for adjusting the
weights which finally leads to different synchronization times for partners and attackers.
However, there are more effects, which show that the two-way communication between A and B
makes attacking the MLP protocol more difficult than simple learning of examples. These
confirm that the security of MLP key generation is based on the bidirectional interaction of the
partners. Each partener uses a seperate, but identical pseudo random number generator. As these
devices are initialized with a secret seed state shared by A and B. They produce exactly the same
sequence of input bits. Whereas attacker does not know this secret seed state. By increasing
synaptic depth average synchronize time will be increased by polynomial time. But success
probability of attacker will be drop exponentially Synchonization by mutual learning is much
faster than learning by adopting to example generated by other network. Unidirectional learning
and bidirectional synchronization. As E can’t influence A and B at the time they stop transmit due
to synchrnization. Only one weight get changed where, = T. So, difficult to find weight for
attacker to know the actual weight without knowing internal representation it has to guess.
11. FUTURE SCOPE CONCLUSION
i s
i s
This paper presented a novel approach for generation of secret key proposed algorithm using
MLP simulation. This technique enhances the security features of the key exchange algorithm by
increasing of the synaptic depth L of the MLP. Here two partners A and B do not have to
exchange a common secret key over a public channel but use their indistinguishable weights or
outputs of the activated hidden layer as a secret key needed for encryption or decryption. So
likelihood of attack proposed technique is much lesser than the simple key exchange algorithm.
Future scope of this technique is that this MLP model can be used in wireless communication.
Some evolutionary algorithm can be incorporated with this MLP model to get well distributed
weight vector.
ACKNOWLEDGEMENT
The author expresses deep sense of gratitude to the DST, Govt. of India, for financial assistance
through INSPIRE Fellowship leading for a PhD work under which this work has been carried out.
38. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
15
REFERENCES
[1] Atul Kahate, Cryptography and Network Security, 2003, Tata McGraw-Hill publishing Company
Limited, Eighth reprint 2006.
[2] Sarkar Arindam, Mandal J. K, “Artificial Neural Network Guided Secured Communication
Techniques: A Practical Approach” LAP Lambert Academic Publishing ( 2012-06-04 ), ISBN:
978-3-659-11991-0, 2012
[3] Sarkar Arindam, Karforma S, Mandal J. K, “Object Oriented Modeling of IDEA using GA based
Efficient Key Generation for E-Governance Security (OOMIG) ”, International Journal of
Distributed and Parallel Systems (IJDPS) Vol.3, No.2, March 2012, DOI :
10.5121/ijdps.2012.3215, ISSN : 0976 - 9757 [Online] ; 2229 - 3957 [Print]. Indexed by: EBSCO,
DOAJ, NASA, Google Scholar, INSPEC and WorldCat, 2011.
[4] Mandal J. K., Sarkar Arindam, “Neural Session Key based Traingularized Encryption for Online
Wireless Communication (NSKTE)”, 2nd National Conference on Computing and Systems,
(NaCCS 2012), March 15-16, 2012, Department of Computer Science, The University of
Burdwan, Golapbag North, Burdwan –713104, West Bengal, India. ISBN 978- 93-808131-8-9,
2012.
[5] Mandal J. K., Sarkar Arindam, “Neural Weight Session Key based Encryption for Online
Wireless Communication (NWSKE)”, Research and Higher Education in Computer Science and
Information Technology, (RHECSIT- 2012) ,February 21-22, 2012, Department of Computer
Science, Sammilani Mahavidyalaya, Kolkata , West Bengal, India. ISBN 978-81- 923820-0-
5,2012
[6] Mandal J. K., Sarkar Arindam, “An Adaptive Genetic Key Based Neural Encryption For Online
Wireless Communication (AGKNE)”, International Conference on Recent Trends In Information
Systems (RETIS 2011) BY IEEE, 21-23 December 2011, Jadavpur University, Kolkata, India.
ISBN 978-1-4577-0791-9, 2011
[7] Mandal J. K., Sarkar Arindam, “An Adaptive Neural Network Guided Secret Key Based
Encryption Through Recursive Positional Modulo-2 Substitution For Online Wireless
Communication (ANNRPMS)”, International Conference on Recent Trends In Information
Technology (ICRTIT 2011) BY IEEE, 3-5 June 2011, Madras Institute of Technology, Anna
University, Chennai, Tamil Nadu, India. 978-1-4577-0590-8/11, 2011
[8] Mandal J. K., Sarkar Arindam, “An Adaptive Neural Network Guided Random Block Length
Based Cryptosystem (ANNRBLC)”, 2nd International Conference on Wireless Communications,
Vehicular Technology, Information Theory And Aerospace Electronic System Technology”
(Wireless Vitae 2011) By IEEE Societies, February 28- March 03, 2011,Chennai, Tamil Nadu,
India. ISBN 978-87-92329-61-5, 2011
[9] Mandal J. K., Sarkar Arindam, “Neural Network Guided Secret Key based Encryption through
Cascading Chaining of Recursive Positional Substitution of Prime Non-Prime (NNSKECC)”,
International Confference on Computing and Systems, ICCS – 2010, 19–20 November,
2010,Department of Computer Science, The University of Burdwan, Golapbag North, Burdwan –
713104, West Bengal, India.ISBN 93-80813-01-5, 2010
[10] R. Mislovaty, Y. Perchenok, I. Kanter, and W. Kinzel. Secure key-exchange protocol with an
absence of injective functions. Phys. Rev. E, 66:066102,2002.
[11] A. Ruttor, W. Kinzel, R. Naeh, and I. Kanter. Genetic attack on neural cryptography. Phys. Rev. E,
73(3):036121, 2006.
[12] A. Engel and C. Van den Broeck. Statistical Mechanics of Learning. Cambridge University Press,
Cambridge, 2001.
[13] T. Godhavari, N. R. Alainelu and R. Soundararajan “Cryptography Using Neural Network ” IEEE
Indicon 2005 Conference, Chennai, India, 11-13 Dec. 2005.gg
[14] Wolfgang Kinzel and ldo Kanter, Interacting neural networks and cryptography, Advances in
Solid State Physics, Ed. by B. Kramer (Springer, Berlin. 2002), Vol. 42, p. 383 arXiv- cond-mat/
0203011, 2002
[15] Wolfgang Kinzel and ldo Kanter, Neural cryptography proceedings of the 9th international
conference on Neural Information processing(ICONIP 02).h
[16] Dong Hu A new service based computing security model with neural
cryptographyIEEE07/2009.J
39. International Journal of Mobile Network Communications Telematics ( IJMNCT) Vol. 4, No.4,August 2014
16
Arindam Sarkar
INSPIRE FELLOW (DST, Govt. of India), MCA (VISVA BHARATI, Santiniketan,
University First Class First Rank Holder), M.Tech (CSE, K.U, University First Class
First Rank Holder). Total number of publications 25.
Jyotsna Kumar Mandal
M. Tech.(Computer Science, University of Calcutta), Ph.D.(Engg., Jadavpur
University) in the field of Data Compression and Error Correction Techniques,
Professor in Computer Science and Engineering, University of Kalyani, India. Life
Member of Computer Society of India since 1992 and life member of cryptology
Research Society of India. Dean Faculty of Engineering, Technology Management, working in the
field of Network Security, Steganography, Remote Sensing GIS Application, Image Processing. 25
years of teaching and research experiences. Eight Scholars awarded Ph.D. and 8 are pursuing. Total
number of publications 267.